KR20210043585A - High temperature infrared radiator element and method - Google Patents

Abstract

An IR emitter element 1 suitable for use as a small infrared emitter (micro hot plate) in a gas sensor, IR spectrometer or electron microscope is provided. The micro hot plate comprises a plate 2 supported by a number of support arms 4. The plate and arm are fabricated from a MEMS device comprising a single continuous piece of electrically conductive refractory ceramic such as hafnium carbide (HfC) or tantalum hafnium carbide (TaHfC). Each arm 4 functions as a heating element for the plate 2 in addition to providing structural cantilever support for the plate 2. The plate 2 is heated by applying a voltage across the arm 4. The arm 4 can be shaped to absorb the thermodynamic stress that occurs during heating and cooling of the arm and plate. For example, ^{a plate that may have an area of less than 0.05 mm 2} and a thickness of 1% to 10% of the maximum dimension of the plate 2 may be heated to 4,000 K or more with a duty cycle of less than 0.5 ms and cooled again, so that It is possible to allow pulsed operation with frequencies of up to 2 kHz. The small size (10-200 μm) and low power consumption (eg 10-100 mW) make the micro hot plate suitable for use in cryogenic applications, small devices or battery powered devices such as cell phones.

Description

High temperature infrared radiator element and method

The present invention relates to the field of microelectromechanical systems (MEMS) hotplates, such as can be used, for example, as IR emitters or IR spectrometers for gas sensors. The present invention is not particularly exclusive, but is a radiator element for an IR emitter capable of emitting broadband infrared radiation at temperatures above 2000 K or above 3000 K, which can be pulsed at frequencies above 1 kHz or even 2 kHz, for example radiator element).

MEMS micro hot plates are available in the near-infrared or mid-infrared spectral range for a variety of applications, such as infrared spectroscopy, lighting for gas detection, hot plates for chemical platforms, or hot plate inserts in transmission electron microscopy (TEM) or scanning electron microscopy (SEM). It is used as an infrared emitter. Currently known devices can operate at pulse rates of up to about 100 Hz. There is a need to increase the operating frequency of these devices or reduce power consumption without compromising the IR emission characteristics.

The published patent applications EP2848914A1 and W02013183203A1 describe an emitter device in which a resistive layer is formed on an insulator. Heat is generated in the resistive layer, and the metal connector provides electrical energy without excessive heat generation to the plate. In each case, the fabrication of the device is relatively complex, and the emission properties of the emitter are limited by several conflicting constraints on the properties of the resistive layer, including sheet resistance, emissivity and melting point.

US 6297511 describes a MEMS IR emitter, a suspended membrane in which the emitter element is a sandwich of a resistive conductive layer between two insulating layers.

Journal of Micromechanics and Microengineering, vol. The article "High-Temperature MEMS Heater Using Suspended Silicon Structures", published in 19 (2009) 115011 (8pp), describes a MEMS hot plate in which a suspended silicon beam is superimposed with platinum tracks that act as resistance heaters. The device described in this article can operate at temperatures up to 1,300K with a pulse rate of 100Hz. U.S. patent US7968848 describes a similar heater with suspended silicon emitter elements.

In a contribution entitled "Molybdenum MEMS Micro Hot Plates for High Temperature Operation" by L. Mele et al, published in Sensors and Actuators A 188(2012) pp 173-180 published by Elsevier, molybdenum as a heater filament material is used as a silicon nitride insulator. It is suggested to sputter the layer. Although molybdenum has a high melting point (2966K), the operating temperature of the device is limited by other materials used, such as silicon and silicon nitride, which degrade at temperatures up to 1,600K and 2,000K, respectively. In theory, an effective thermal IR emitter can operate at temperatures above about 1,300K, but for practical implementation reasons, it is rated at a lower temperature (e.g., up to 1,000K). This upper operating temperature limit imposes a limit on the intensity and bandwidth of infrared radiation that can be generated for a given power consumption.

It is also known to provide a source of metal atoms by coating a silicon plate with a metal (eg, silver) and then heating the plate to evaporate the metal with heat generated by the resistive support spring of the plate. This technique is described in a contribution by Han Han et al. "Programmable Source of Solid State Atoms for Nanomanufacturing" published in Nanoscale, 2015, 7, 10735. The MEMS evaporator comprises a polysilicon plate supported by two polysilicon springs that act as resistive heating elements when an electric current is applied. The metal is deposited on the polysilicon plate, and the plate is protected from eutectic interactions with the metal by an _{Al 2} 0 _{3 thin film coating.} The heated plate provides a source of metal atoms used in, for example, nanolithography. The deposited metal also serves to distribute heat from the two springs across the plate. This document does not propose to apply a metal-coated plate for use as a micro hot plate IR emitter.

Prior art macroscopic halogen sources capable of generating IR at 3,000K can be used, but these are large, high-power, and cumbersome devices.

The present invention aims to overcome at least some of the shortcomings of the prior art. In particular, there is a need for radiator devices that have a smaller heat capacity and/or can be manufactured in a smaller area for a given radiant power than prior art radiators that can operate at higher frequencies. To this end, the radiator device according to the present invention is described in the appended claim 1, the infrared ray generating method according to the present invention is described in claim 14, and the manufacturing method according to the present invention is described in claim 16. Further variations of the invention are described in the dependent claims.

Using materials that can significantly raise the upper operating temperature limit can significantly increase the intensity of the emitted radiation, as explained by Stefan-Boltzmann's law for total IR power and Planck's law for spectral density. Thus, it is possible to greatly reduce the area of the radiator plate for a given radiant power, reduce power consumption, and/or increase the maximum heating/cooling frequency. The temperature of the emitter defines the peak spectral wavelength. In this case the peak at 1000K could be in the IR at about 3 microns, for example. When the plate is heated to 4000K, the peak is in the visible range of the electromagnetic spectrum. Exceeding the temperatures obtained by conventional halogen lamps, but using a MEMS form factor and scalable manufacturing method, the radiating element can be constructed.

The invention will be described in detail with reference to the accompanying drawings. In the drawing:
1 and 2 each illustrate an isometric view and a plan view of a first simplified schematic diagram of a basic radiator plate according to the invention.
3 illustrates an isometric view of the radiator of FIGS. 1 and 2 formed over the concave recessed area of the substrate in the first exemplary mounting configuration of the radiator device according to the present invention.
4 and 5 illustrate schematic cross-sectional views of first and second variants of the concave substrate, taken along the AA axis of FIG. 3.
6 illustrates a schematic plan view of a second exemplary mounting configuration of the radiator plate of FIGS. 1 and 2;
7 illustrates a schematic plan view of a third exemplary mounting configuration of a radiator plate according to the present invention.
8 illustrates a schematic plan view of a modified example of the radiator according to the invention in which the heater/support arm is tapered.
9 is a schematic diagram of another variant of the radiator device according to the invention in which the heater/support arm is formed as an elastic element having a first exemplary configuration for absorbing thermodynamic expansion, contraction or other motion of the radiator plate or arm. Illustrate the plan view.
10 is a schematic plan view of another variant of the radiator device according to the invention wherein the heater/support arm is formed as an elastic element having a second exemplary configuration for absorbing thermodynamic expansion, contraction or other motion of the radiator plate. Illustrate.
11 is a schematic plan view of another variant of the radiator device according to the invention wherein the heater/support arm is formed as an elastic element having a third exemplary configuration for absorbing thermodynamic expansion, contraction or other motion of the radiator plate. Illustrate.
12 and 13 respectively illustrate an isometric view and a plan view of another modified example of the radiator device according to the present invention.
14 illustrates a schematic plan view of another variant of the radiator device according to the invention wherein the heater/support arm is formed as an elastic element having another exemplary configuration for absorbing thermodynamic expansion, contraction or other motion of the radiator plate. .
15 illustrates another variant of the radiator according to the invention in which a plurality of support arms are connected to a common mounting. The figure also shows how the radiator device of the present invention can be driven by a PWM signal, a voltage source or a current source.
16 shows an example of an equivalent resistance circuit for the modified example of the radiator illustrated in FIG. 15. The connector 3 is, for example, at or close to a set potential or a ground potential. 8 arm (4) each of which is represented by resistance than is usually much larger, preferably at least 10 times, more preferably at least 100 times greater resistance (R ₄₎ of the plate (2), exemplified by R ₂ .
17 shows an example of the heat flow presented for the exemplary modification shown in FIG. 15. The power is converted into thermal heat, preferably equally to all eight arms 4. This heat flows to the plate (2). Some heat is radiated from the plate (eg infrared) as radiation. The highest hot spot of the radiator device on the arm 4 is indicated by a dotted line marked _{T max.}
18 shows an example of an array configuration of a multiple radiator device according to the present invention.
Figure 19a shows a graph of the temperature distribution across the emitter device according to the invention at different temperatures. The dotted line represents the boundary of the element comprising the connecting pad 3, the arm 4 and the plate 2.
Figure 19b shows a graph of the intensity distribution of radiated radiation across the emitter device according to the invention at different temperatures. The dotted line represents the boundary of the element comprising the connecting pad 3, the arm 4 and the plate 2.
19C shows a graph of the temperature gradient between the connection pad 3, the arm 4 and the plate 2. The dotted line represents the boundary of the element comprising the connecting pad 3, the arm 4 and the plate 2.
20 shows a graph of _{spectral radiation (R spec} ) versus emission wavelength at various temperatures.
21 illustrates a graph of displacement and temperature versus time in an exemplary radiator device in accordance with the present invention.
22-25 shows a variation of the configuration of a micro-hot plate IR emitter including an emitter according to the present invention.
26A-26G show an exemplary manufacturing process for manufacturing an IR emitter comprising an emitter according to the present invention.
26H illustrates a modification of the manufacturing wafer of the present invention in which an application device is manufactured on the IR emitter illustrated in FIG. 26G.
27 shows a first example of a Fabry-Perot IR spectrometer application apparatus including an emitter plate according to the present invention.
28 shows a second example of a Fabry-Perot IR spectrometer application apparatus including a radiator according to the present invention.
It should be noted that the drawings are provided merely to aid in understanding the principles underlying the present invention, and should not be regarded as limiting the scope of protection sought. Where the same reference numerals are used in different drawings, they are intended to indicate similar or identical features. However, it should not be assumed that the use of different reference numbers is intended to specify the degree of difference between the referenced features.

As used herein, the term “conductive refractory material” refers to an electrical resistance of less than 1 Ohm cm and a thermal conductivity in the range of 10 W/mK to 2200 W/mK, and at very high temperatures (ie, greater than 1,600 K, without degradation or change to a significant extent It is used to refer to an inorganic material that can be heated (for a short period of at least, e.g., 10 ms), preferably to a temperature of more than 2,000 K, more preferably more than 2,500 K, or more preferably more than 3,000 K. . The Young's modulus of the refractory material can be in the range of 100-1000 GPa, and the flexural strength can exceed 100 MPa. All property values were obtained at room temperature unless otherwise specified. Suitable materials include carbon allotropes such as graphite, diamond (doped), graphene, fullerene, carbon nanotubes (CNT) and fullerenes or CNT deposits (single crystal or fullerene metal complex); Or hafnium carbide, tantalum carbide, tantalum hafnium carbide, tungsten carbide, titanium carbide, niobium carbide, hafnium boride, hafnium nitride, or And electrically conductive refractory ceramics such as hafnium tantalum nitride, or any partial combination of the aforementioned compounds (eg Hf _{0.98 C).}

The term "micro-hotplate" is used herein to refer to a small component that provides a very small and very hot surface and/or a very small and very strong infrared radiation source (IR emitter). The terms “radiator”, “radiator device” and “radiator element” refer to very small plate-like or other structures that are integrated into a micro hot plate device such as an IR emitter device and heated to produce the necessary intense infrared light. The area of the emitter plate ^{may be less than 0.1 mm 2} or less than 0.05 mm ² , for example.

The invention has been described in connection with its application to the generation of broadband infrared radiation. However, the emitter device of the present invention can be used to emit light in the visible spectrum in addition to or as an alternative to IR. Applying the principles of the invention to the generation of visible light can be considered a separate invention.

1 and 2 show a first example of a radiator 1 suitable for use as a radiator element of an IR emitter. Unless otherwise specified, the principles and features described in connection with this example apply to the other examples described below and in general to other variations of the invention. The radiator 1 comprises a structure 2 (a square, rectangular or parallelogram plate in this example) formed of a conductive refractory material as described above. As will be described, the material of the radiator is partially on the substrate (pad 3) and later removed on the sacrificial material (plate 2 and arm 4) using, for example, known semiconductor manufacturing techniques. ) Can be grown or deposited. It has been found through testing and/or simulation that hafnium carbide exhibits suitable thermodynamic properties (eg, sufficient thermal and electrical conductivity, thermal and chemical stability, and adequate elasticity) for use in the manufacture of the emitter element 1.

For operation in the above-mentioned high temperature range, the radiator 1 can advantageously be held in a vacuum or in a lean inert gas. The plate 2 may have a size of 2 to 500 μm (the length of one side of a square/rectangle/parallelogram in this example), for example, 0.1 of the side length of a square (or the maximum dimension of the plate if not square). It may have a thickness of to 10%. The structure is illustrated as a continuous planar piece formed of a material of uniform thickness, but the structure 2 is alternatively a non-planar (e.g., concave) shape and/or a lattice or ladder or serpentine configuration. It can be formed with different topologies or perforations or surface features or textures. Such surface features or textures can be added to enhance emission from the surface, for example. While a square plate is illustrated in Figures 1-3, in practice the plate can be of any convenient shape. For example, it may be circular, triangular, square or polygonal. The plate 2 and arm 4 may be coplanar or, as illustrated, may not be coplanar. The arm 4 may be formed such that the plate 2 is vertically offset from the plane of the pad 3 so that the plate rises above the height of the substrate 8 when the pad contacts the substrate 8. The arm is represented as having some displacement. This may be necessary to absorb the thermodynamic displacement during heating and cooling. However, this example is only intended to facilitate understanding of the present invention. Indeed, as discussed below, arms 4 of other shapes and configurations may be required to provide adequate motion absorption.

The plate 2 is formed by means of a support element or spring extending between the plate 2 and the connection pad 3 or multiple arms 4, also referred to as heaters or heater springs (for example, four as in the example presented). Is supported. As explained in connection with FIG. 3, the connection pad 3 is designed to provide a mechanical connection to the substrate such that the plate is supported against the substrate only by the arm 4 and the pad 3. The connection pad 3 provides an electrical connection to the arm 4 and thus to the plate 2. The plate 2, the pad 3 and the arm 4 are preferably formed from one continuous piece of material. The material may have a uniform bulk resistance, or may have a bulk resistance different from other parts of the radiator 1. For example, the material of arm 4 may be doped or treated to provide the arm with a lower bulk resistance than the material of plate 2. For example, the bulk resistance of the material of the arm can preferably be in the ^{range of 10 -5} to 0.1 Ohm cm, for example. In the illustrated example of FIG. 1 each arm 4 has a length 5, a width 6 and a thickness 7 and has a cross-sectional area much smaller than that of the plate 2. Width 6 and thickness 7 advantageously provide an aspect ratio (thickness:width) of 1:10 to 1:1 for improved stability. By way of example, the arm length may be, for example, 5 to 200 μm, preferably 10 to 150 μm, or more preferably 15 to 80 μm. The thickness may be, for example, 0.1 to 5 μm, or more preferably 1 to 3 μm. The width of the narrowest part of the arm may be, for example, 0.1 μm or less or 20 μm or more. In the case of a single layer material such as graphene, the plate 2 and arm 4 may be as thick as only a single layer (or a stack of single layers when multiple layers are stacked).

When a voltage is applied between the two ends of the arm 4 (eg, through an electrical connection to the connection pad 3), heat is generated by the ohmic heating of the arm 4. The arm 4 is thermally connected to the radiator plate 2. The plate 2, which can be held in a vacuum state, acts as a heat sink for the heat generated in the arm 4. At higher temperatures (eg 3000 K to 4000 K), a portion of the heat generated in arm 4 is radiated from plate 2 as IR increases.

3 shows how the radiator 1 of FIGS. 1 and 2 can be mounted or fabricated on a substrate 8 with a concave recess 9 underneath it. Such structures can be formed using known semiconductor manufacturing techniques, for example. As discussed below, the concave recess 9 may be formed as a flat curved (e.g., parabolic) surface as illustrated in FIG. 4, or may be formed in another shape, such as an inclined wall illustrated in FIG. 5. . The concave recess can improve the net radiant power of the micro-hot plate assembly by having the function of inducing IR radiation outward from the plate 2 (away from the substrate 8). This can be achieved by means of a reflective surface or coating 15.

In the example shown in Figures 1-5 there are 4 heater arms/springs. However, the number of springs (arms 4) may be 2-16 or more. Since this makes it easier to balance the input and output currents to make the heat distribution more uniform, the number can be the same. The heat generated in the arm is conducted to the plate 2 as mentioned. Experiments and simulations show that the plates of this example can reach a thermal steady state within 0.2 to 0.5 ms, which means that the temperature can be controlled between 2 kHz and 5 kHz. If the plate 2 is made smaller, an operating frequency of 50 kHz or even 100 kHz is possible.

The total power consumption of the plate can be, for example, a minimum of 10 mW or a maximum of 1 W per plate, depending on the emitter configuration and plate size.

The mechanical resonance is about 1 MHz, which means that the device is insensitive to all common external vibrations and shocks. This value can be lower or higher depending on the shape of the device and the material of construction chosen. The mechanical mode can be more than 1 kHz, preferably more than 10 kHz or more preferably more than 100 kHz, so that the structure can withstand impacts due to collisions, for example when the device is dropped.

6 illustrates an alternative configuration for mounting the radiator 1 to the substrate 8. In this case, the substrate 8 may be coated or polished with a reflective surface or layer 15, and the pad 3 is formed on the spacer 23, which serves to keep the plate 2 away from the substrate 8 Or mounted. The arm 4 is symbolically represented by a wavy line, but this marking is intended to correspond to any variation of the heater/spring arm that can be used in general. The spacer 23 can be formed of the same material as the sacrificial layer used to separate the device material from the substrate 8, or alternatively the spacer is preferably formed of a separate insulating or conductive material with high thermal conductivity. Can be.

7 shows another variant in which, instead of the spacer 23 of FIG. 6, the pad 3 is shaped to provide the desired vertical offset from the substrate 8 and the reflective surface/layer 15.

8 shows a variation of the radiator 1 that increases as the resistance of the arm 4 is closer to the plate 2. This reduces the bulk resistance of the material of the arm 4 by tapering the cross-sectional area of the arm 4 (e.g., by tapering the width 6 as illustrated) and/or with variable doping of the material of the arm 4. This can be achieved by increasing toward the point of intersection with the plate 2. A local increase in resistance means a local increase in ohmic heating near the plate, so more of the heat generated is transferred to the plate. This tapering of resistance can be used in any modification of the radiator of the present invention.

9-11 show different variants of the radiator element 1 in which the arm 4 is configured to absorb mechanical deformation of the plate 2 and/or arm 4 when heated and cooled. The coefficient of thermal expansion of refractory ceramics such as HfC ^{can be in the range of 1-10 × 10 -6} /K, and by shaping the heater/support spring 4 in a manner that provides an elastic buffer for expansion and contraction operation, such thermodynamics It is desirable to allow the behavior. At a temperature change of 3,500K, the plate 2 can expand or contract by about 1-3% (over the entire plate). Figure 9 shows a variant with a larger (e.g. 100 μm wide) octagonal plate 2, while Figure 10 shows a variant with a smaller octagonal plate 2 (e.g. 10-30 μm wide). Show In the example of two octagonal plates, each connection pad 3 is connected to the plate 2 via two arms 4. In the example of a small plate (Fig. 10), the thermodynamic deformation of the plate 2 is minimized and the temperature is almost uniform throughout the plate.

11 shows another variant in which each pad 3 is connected to the plate 2 by two arms 4. Six pads 3 and six double arms 4 are provided, and the plate 2 has a substantially circular shape. Small holes (circular in this case, but other shapes such as square or rectangular) can be used, for example, to be cut into the plate 2. Depending on the manufacturing process used, these holes can aid in the release phase when the material under the plate is removed to form a freestanding structure. Such apertures may be beneficial to any of the examples provided and are not limited to the embodiment represented in FIG. 11.

In the example of FIGS. 9, 10 and 11, each connection pad 3 is connected to a plurality of head arms 4. In contrast, FIGS. 12 and 13 show a modified example in which each arm 4 is connected to one connection pad 3. In this example, the plate 2 may have a diameter of 10 μm to 200 μm, for example. In this case, the arm 4 and the plate 2 are arranged such that the plate 2 is rotated due to the thermodynamic displacement.

To achieve this with the minimum stress of the arm 4, the arm is shaped to be oriented along a peripheral path parallel to the rotation of the circumference of the plate 2. The arms 4 may optionally include stepped or offset sections such that they overlap so that the longer arms can fit in a small peripheral area around the plate 2. The arm 4 of this exemplary variant is connected to the plate by means of a narrowed radial portion, which is formed, for example, by forming a notch around the plate 2 on both sides of the radial portion. Can be. When voltage is applied across the pad 3, the narrowed radial portion serves as the main heating source of the plate 2.

14 and 15 show a simplified schematic example of a radiator element similar in principle to the variant of FIGS. 12 and 13, wherein the plate 2 is substantially circular and the tangential spring/arm is the plate 2 and the mounting pad (3) extends between. In the example of FIG. 14, each connection pad 3 is connected to the plate 2 by means of four springs 4. In general, it is advantageous to provide an even number of arms/springs as it facilitates the manufacture of a radiator in which the electrical input and output connections are balanced. The more the shorter arm is used, the more uniform the heat distribution of the plate and the faster response time, but the higher the power consumption. Larger plates generally require more springs for effective heating. Longer springs provide better mechanical strain relief but slower response times. Longer and thinner springs are more susceptible to breakage when the device is impacted. The circular plate and tangential spring of FIGS. 14 and 15 generate very little rotational motion when the plate 2 is heated and cooled. This rotation of the device of the example shown in Figs. 12, 13, 14 and 15 contrasts with the previous device example in which no rotation is actually observed because the plate is only radially expanding.

The mechanical stress generated by thermal expansion is relieved by the arm 4. The thermal expansion of the plate 2 and arm 4 should preferably be compensated to ensure mechanical integrity and maintain mechanical, electrical and thermal continuity. In the example of Figures 9, 10 and 11, the thermal expansion causes the arm 4 to bend as the arm 4 extends and the plate 2 expands in the radial direction. In the example given in Figs. 12, 13, 14 and 15, the plate 2 rotates due to the radial expansion of the plate 2 and the linear expansion of the arm 4. The arm 4 will also bend. In the example of Figure 1, thermal stress causes buckling above a given temperature. The arm 4 may have additional structures such as, but not limited to, notched, tapered, and meandering structures to adjust and control mechanical deformation along with electrical and thermal conductivity.

15 also illustrates a pulse width modulated (PWM) signal 10 and a radiator 1 connected to a current source or a biasing voltage source. When driven in PWM mode, the PWM duty cycle can be varied to change the temperature and spectral characteristics of the IR emitted from the emitter. In this example, the drive signal is defined via a PWM signal. The PWM frequency should preferably be at least 10 times higher than the thermal response time of the aforementioned device. In addition to or alternatively to PWM control, the device is typically on the order of 1 V, but can be driven with a DC or AC voltage bias as low as 1 mV or as high as 100 V depending on the emitter 1 shape and material. Voltage biasing will provide stable operation for materials that increase their resistance to temperature. Alternatively, the radiator can be driven by current bias even in DC or AC mode. This current bias is typically in the range of up to 10 mA, but can be higher or lower depending on the emitter 1 shape or the material used. A current biased radiator 1 would be a reliable option for a device in which the arm 4 resistance decreases with increasing temperature.

For example, an additional electrode may be provided to the radiator illustrated in FIG. 15 to measure a voltage between the two pads 3. This allows the resistance of the device to be monitored in real time. The resistance of the plate 2 as well as the arm 4 changes with temperature. By providing an additional electrode as illustrated in FIG. 15, voltage and current can be recorded. From this, the resistance can be determined, and with appropriate correction, the resistance is converted into an IR spectrum and intensity. This feedback can be used to improve the stability of the emitted IR radiation.

16 illustrates an equivalent circuit diagram for the modified example illustrated in FIG. 15. In this variant there are a plurality of arms 4 (8 are illustrated) connected to the plate 2. One or more arms 4 of the first subset (four arms of the first subset in this example) are electrically connected to the first connecting pad 3 on the left (in this example in parallel), and the second sub The arms of the set (four arms of the second subset in this example) are similarly connected in parallel to the second connection pad 3 on the right. If the first pad 3 is set to a potential (V) or close to the potential (V) and the second pad 3 is set to another potential, e.g., a ground potential or close to the ground, the first pad 3 The current (I) is drawn through a subset of paralleled arms (4), through a plate (2), and through a second subset of paralleled arms (4) to a second pad (3) with a lower potential. Flow.

If all arms 4 are electrically equal, the current in each arm 4 is 1/4 as explained by Kirchhoff's law. Consequently, the power consumed by each arm 4 is also the same and _{has a value of R 4} I ₂ /16, where R ₄ is the electrical resistance of each arm 4. Preferably, the resistance R ₂ of the plate 2 is significantly lower than the resistance of each parallel connected subset, and much lower than _{the resistance R 4 of each individual arm 4.} This allows most of the power to be dissipated in the arm 4 so that most of the heat generated by the current flowing through the pad-arms-plate-arms-pad series connection configuration is generated in the arm and less generated in the plate. When viewed in a series-connected circuit of pad-arm(s)-plate-arm(s)-pad, the resistance of each arm subset is preferably at least 10 times, more preferably at least 50 times or even the resistance of the plate. It is at least 100 times or more. In order to minimize the electrostatic interaction between the plate 2 and the substrate 8 or the reflective surface 15, the potential applied to a given pad is equal to the plate potential 2 with the substrate 8 or the reflective surface 15. It can be set to be at the same potential. For example, if one pad is at V/2 potential and the opposite pad is at -V/2, then the potential of plate 2 is preferably zero, which is the potential of substrate 8 and reflective surface 15.

In FIG. 17, the generated power and heat flow is schematically illustrated for the variant illustrated in FIG. 15. The power P _el is converted into thermal energy in arm 4, making arm 4 the hottest point of the radiator. Preferably the highest hot spot of each arm 4 is close to the plate 2 as indicated by _{T max in FIG. 17.} This means that the thermal resistance of the thermal path for the pad 3 is high compared to the thermal resistance for the plate 2. A small amount of the heat emitted by the arm 4 _{flows to the pad 3 as thermal energy P Bth} , where the pad is preferably at a temperature close to the temperature of the substrate, which may be close to the ambient temperature. The ambient temperature may be room temperature (typically 300K), but may be much lower or higher if the emitter is operated in a cryostat or heated environment. Some, preferably most, more preferably substantially all of the thermal energy generated in the arm _{flows as P Plth} to the plate 2, where the plate 2 is heated to the desired temperature. The thermal energy for the plate 2 is radiated as electromagnetic radiation (P _IR-radiation ), for example infrared radiation with some visible light radiation.

As the radiated heat energy is ^{adjusted to ~T 4} , while the thermal conductivity from plate 2 and arm 4 to pad 3 is adjusted to ~T, the emitter becomes more efficient at higher temperatures.

Fig. 18 shows an IR emitter comprising an array of multiple (16 in this example) radiator elements 1 connected so that each can be driven by a separate PWM signal. Such an array can be used, for example, to generate infrared radiation with a specific spectral profile. A substantially larger surface area increases the IR radiation produced. By selecting the number of radiators 1 to be activated, the intensity can be adjusted without shifting the spectrum. A single power supply can drive all radiators in the array with a transistor that sets a PWM or voltage or current signal to each element of the array (emitter 1).

19A shows an example of how the temperature of the plate can vary across the emitter plate for different applied voltages (eg, applied along the AA axis in FIG. 3). The x-axis represents the distance measured from the center point of the axis. The y-axis shows the temperature of the plate 2 and arm 4 measured at corresponding positions along the axis. Part of the curve is marked to indicate which part (pad 3, arm 4 or plate 2) it corresponds to. In this example, the plate has a diameter or transverse dimension of about 70 μm and an arm length of about 15 μm. In each of the three temperature curves illustrated, the temperature of the connection pad 3 is substantially constant at ambient temperature (e.g. 290-300K), but there is a steep temperature gradient in the arm 4 and somewhat flat across the plate 2. There is a temperature profile. For an applied voltage of 1.5 V (in this example), the central region of the plate 2 is colder than the peripheral region near the inner end of the heating/support arm where heat transfer from the arm to the plate takes place. This is due to the result of much greater radiation emitted at higher temperatures (eg about 4,000 K) than at lower temperatures (and thus greater light output and greater cooling of the plate 2). For example, the light power density at 4000K may correspond to ^{about 12 MWm -2 at 0.85 emissivity.} At 4000K, the spectral radiation at a wavelength of 3 microns is about 2 times greater than at 1000K and the peak spectral radiation is about 3 times greater.

The curve illustrated in Fig. 19A corresponds to the curve predicted by Planck's law. On the other hand, FIG. 19B shows how the radiation intensity distribution can actually change with temperature across the same emitter device. Increasing the applied voltage from 0.5 V to 1.5 V (thus increasing the temperature from about 1600 K to more than 4,000 K) increases the radiation intensity more than twice.

FIG. 19C shows an example of how the temperature gradient of temperature can vary across the radiator device during operation as used in FIGS. 19A and 19B. In this example, the temperature gradient has a maximum value at a point in the outer region of the arm 4 adjacent to the pad 3. This is partly because the pad 3 is at a much lower temperature than the arm 4 and partly because of the small cross-sectional area (and thus high resistance) of the arm at that point. The combination of these two factors means that the maximum temperature gradient is provided at that point, but the minimum thermal energy flow is provided. At the other end of the arm 4 on the side of the plate 2, the temperature gradient is minimal because the plate 2 is at or near the desired radiation temperature. In this state, the temperature gradient is minimal, the heat energy flow from the arm to the plate is maximum, and the heat energy released from the plate 2 is supplemented by the heat inflow of the heated arm 4. As mentioned elsewhere in this description, most or almost all of the heat to increase the temperature of the plate is generated by the resistive heating of the arm 4, and by the much lower resistance of the plate 2 (relative to the arm). Generates little or no heat.

Figure 20 shows how the spectral radiation (y-axis, arbitrary units) varies with wavelength (x-axis) at different plate temperatures.

Fig. 21 shows that the temperature (left vertical axis) and thermodynamic displacement (right vertical axis) during the heating cycle of the plate 2 of the radiator 1 are changed in time as in the case where the voltage bias is applied as a step function starting at time t = 0. Illustrate how it changes accordingly. As can be seen from the curve, the thermal cycle in this example can reach a steady state within about 0.3 ms. A similar rate can be seen in the cooling phase (not shown).

22 illustrates an IR emitter comprising a radiator element constructed to have an radiator plate 2 in a vacuum-treated package and an IR-transmitting window 14 which also functions as a closure for the vacuum chamber. The directional reflective surface 15 is arranged to reflect the IR generated by the plate 2 towards the window 14. In all variants of the invention, the reflective surface or layer 15, if present, can be formed in the recess 9 under the emitter plate 2 and/or on the sidewall of the structure above the emitter plate. The window 14 may be formed of, for example, sapphire, germanium, silicon or diamond. The material of the window 14 can be selected for specific IR filtering properties as needed. The substrate 8 can preferably be made of a material having high thermal conductivity, such as silicon. The spacer layers 11 and 13 are typically electrically insulating and may be formed of, for example, silicon oxide, silicon nitride and/or other dielectrics or insulators, preferably having high thermal conductivity. The material of the spacer layers 11 and 13 can be formed of a material such as a semiconductor and a conductor such as silicon by means of an appropriate electrical insulating barrier added to the interface of the spacer layers 11 and 13. The reflective wall may be formed by anisotropic etching, grayscale lithography or otherwise removing the silicon between the 111 and 100 sides, leaving a surface of 125.26° relative to the plane of the substrate 8. The interior of the IR emitter is preferably ^{evacuated to 10 -6} Torr or preferably 10 ^-6 or more.

In addition to sealing the vacuum chamber, the window 14 may optionally perform other functions, such as optical filtering or lensing. The window material can partially define the IR spectrum emitted by the device, and additional coatings can be added to the surface to alter the reflectivity and transmittance properties. 23 illustrates the patterning, molding or other treatment of the window 14 to provide a Fresnel lens on the outer surface to change the focal characteristics of the IR emitter. Window 14 may alternatively or additionally be configured to act as a pass or cut filter for specific spectral components of the IR. The window may be patterned or doped, or a material may be deposited thereon to form a metamaterial surface for wavelength selection and/or light path adjustment.

24 shows an arrangement in which the chamber walls have a parabolic or elliptical profile and the chamber walls are coated with an IR-reflective coating 18, such as a metal layer, to improve the ratio of IR exiting the vacuum chamber through the window 14. Show. The IR-reflective coating 18 can be included in any of the variants described. In a further variant, FIG. 25 shows how the radiator plate 2 can be made steerable or vibrable, for example by means of a capacitive or thermodynamic actuator (not shown). The window can be optionally shaped and/or coated to provide the desired optical effect of IR emission.

The steering configuration illustrated in FIG. 25 can be achieved by dividing the reflective surface 15. By providing an additional electrical lead to access the segmented metal reflector layer 15, an electric potential can be applied between the reflector 15 and the plate 2. Since the plate is held by a spring (arm 4), the plate 2 is free to tilt, tilt or move the piston. The force required to change the plate orientation is given by the capacitive force (approximately F ~ ε ₀ A/d*V ² , where F is the force, ε ₀ is the permittivity of free space, and A and d are the divided electrodes and The area of the segment and separation between the plates 2, V is applied by the divided reflector and the voltage between the plates). Increasing the number of segments affects the tilt/tilt/piston ability of the plate, and four segments are generally sufficient. Mechanical angles of typically 10 degrees can be reached, which can be used to steer the emitted light and/or fine-tune the alignment of the device with other optical components.

26A-26G show a simplified example of a method by which an emitter can be fabricated at a wafer scale along with other components of an IR emitter on a wafer substrate 8. In Fig. 26B, the cavity 9 is etched and a reflective (eg, metal) coating 15 is deposited. Then, a sacrificial material 20 is added, planarized and patterned, on which an element layer 21 (high temperature ceramic) is formed, and the pad 3, arm 4 and plate 2 constituting the element are formed. It is patterned. In Fig. 26E, the sacrificial material has been removed so that each radiator plate of the element layer 21 remains held above each reflective cavity 9. Then, a window 14 is added using the handle 24 supported by the spacer 23 and bonded by the junction 22 at high vacuum to surround the hot emitter element as illustrated in FIG. 26G. It is possible to form a sealed IR emitter each comprising (27). The top layer (window 14 in this example) can then serve as the basis for an additional wafer-scale manufacturing process. The application device 25 may be fabricated directly on the IR emitter wafer, or it may be fabricated individually and then aligned and bonded to the IR emitter wafer to form an application device having an integrated IR emitter as shown in FIG. 26H. This is useful, for example, to downsize an IR interferometer or similar application device. The individual stacked assemblies (eg gas sensors, chemical sensors, IR interferometers, etc.) can then be diced cut and packaged into individual surface mount technology packages or other individual devices. At this level of miniaturization, it is possible to manufacture finished application devices with IR emitters with a footprint of 300 x 300 µm or smaller.

26A-26G show that the first wafer holds the radiator element and associated electron leads, and the second wafer has a spacer 23, an IR window and an additional functional surface (bonding layer 22 and a getter layer for gas absorption). ), an exemplary method using a two-wafer approach is described. The two wafers are bonded and sealed in a vacuum to form a separate independent vacuum chamber. In an alternative embodiment, the entire stack can be fabricated in a stacked manner.

Continuing in FIG. 26E, after additional spacer material has been added and patterned, a patterned getter material and finally an IR window 14, which may be formed of, for example, germanium may be deposited.

27 and 28 illustrate cross-sectional views of two variants of an application device in which a high-temperature ceramic radiator plate is incorporated. In the example presented, the application device is placed over a concave recess 9 of a vacuum chamber 27 hermetically sealed with an IR transparent window 14 supported on a Fabry-Perot interferometer (FPI) 25 and a spacer 23. It is a gas sensor comprising an IR emitter 1 with a held refractory ceramic radiator plate 2. In addition, in a vacuum, the cavity 9 has a reflective coating or surface 15 for directing the IR generated by the radiator plate 2 toward the window 14. Electrical connection of the radiator to the pad 3 is allowed by the contact 30, so that a voltage or current is applied across the heater/support spring 4, causing the plate 2 to exceed 1,600K, 2,000K or 2,500K. It can be heated to a temperature of up to 4000K or higher. The radiator plate 2, spring 4 and pad 3 can be made of, for example, HfC, TaC, or TaHfC, or some other suitable refractory ceramic material. The FPI device 25 includes upper and lower partial reflectors 31 and 32, which together form a Fabry-Perot resonant chamber and act as an optical band pass filter.

The FPI can be adjusted by moving the upper reflector 32 away or close to the lower reflector by means of an actuator 29 in a known manner. In the variant illustrated in Fig. 27, the IR detector of the FPI is positioned in a fixed position on the inner surface of the upper closure member of the FPI device. In contrast, in the variant of FIG. 28, the FPI IR detector is mounted to move together with the upper reflective element 32 of the FPI filter 25. This arrangement in which the FPI IR mirror is mounted to a moving reflector or mounted to move with a moving reflector is not limited to being used in the reflector device of the present invention and may be used with other types of IR emitters.

In two variants of FPI, when they are implemented as gas/chemical sensors, the gas to be detected/analyzed is within the cavity region 33 comprising a space between the two partial reflecting mirrors 31 and 32.

The FPI 25 may be fabricated directly to the window 14 of the IR emitter 1, or it may be individually fabricated and then aligned and bonded to the window 14 of the IR emitter 1.

Other possible applications of an IR emitter comprising a radiator device according to the invention include a Michelson Morley Interferometer (MMI). In MMI, the optical path is in the plane with the wafer, which allows for a much longer optical path, which is advantageous for gas detection.

Claims (17)

As an emitter device (1) for an IR emitter micro-hot plate,
Said radiator device (1) comprises an IR emitter element (2) and a plurality of support arms (4) connected to said emitter element (2),
The emitter element (2) is suspended by the arm (4),
The emitter element (2) can be heated to a predetermined IR radiation temperature by resistive heating of the arm (4). 2. The predetermined IR radiation according to claim 1, wherein the emitter element (2) is completely or in large part through the heat conduction from the arm (4) into the emitter element (2) by means of the resistive heating of the arm (4). A radiator device configured to be heatable to a temperature. The method according to claim 1 or 2, wherein the IR radiation temperature is higher than 1,600K, preferably higher than 2,000K, more preferably higher than 2,500K, even more preferably higher than 3,000K, or even more preferably higher than 3,500K. The radiator device (1), which is a temperature above K. 4. The radiator device (1) according to one of the preceding claims, wherein the emitter element (2) and the arm (4) are formed from a single continuous piece of material. 5. The radiator device (1) according to claim 4, wherein the material is an electrically conductive refractory ceramic. 6. The radiator device (1) according to claim 5, wherein the ceramic comprises carbon, HfC, TaHfC or tungsten carbide. The radiator device (1) according to one of the preceding claims, wherein the number of arms (4) is an even number, and the even number is at least 4, preferably at least 6, or more preferably at least 8 . 8. The shape and/or size of the emitter element (2) and/or the arm (2) according to one of the preceding claims, wherein the arm (4) is formed during heating and cooling of the emitter element (2). The radiator device (1), resiliently deformable to absorb the thermodynamic changes of. The method according to one of the preceding claims, wherein each of the arms (4) has a cross-section that varies along its length such that its cross-sectional area is minimal in the area of the arm (1) adjacent to the emitter element (2). , Radiator device (1). An IR emitter device comprising a radiator device (1) according to one of the preceding claims, wherein the emitter element (2) and the arm (4) are in a housing comprising an IR-transparent window (21). Encapsulated, IR emitter device. The IR emitter device according to claim 10, wherein the housing is ^{evacuated to less than 10 -3} Torr, ^{less than 10 -4} Torr, preferably less than 10 ^-5 Torr, or more preferably ^{less than 10 -6 Torr.} Gas sensing, pressure sensing, comprising an IR emitter device according to any one of claims 10 and 11, or an IR emitter device comprising a radiator device (1) according to claim 1, Gas analysis, IR spectrometer, SEM or TEM device. A portable communication device comprising a gas sensing, pressure sensing, gas analysis, IR spectrometer, SEM or TEM device according to claim 12. As a method of generating broadband infrared radiation:
Using an IR emitter device according to claim 10 or an IR emitter device comprising a radiator device (1) according to claim 1; And
The emitter element 2 is heated to a temperature above 1,600 K, preferably above 2,000 K, more preferably above 2,500 K, even more preferably above 3,000 K, or even more preferably above 3,500 K. Step of applying voltage across the arm (4)
Including, the method. 15. The method according to claim 14, comprising pulsing the voltage to a frequency greater than 200 Hz, preferably greater than 700 Hz, or more preferably greater than 1,000 Hz. A method of manufacturing an IR emitter device, comprising a first manufacturing process of manufacturing a multiple emitter device (1) according to claim 1 on a single wafer. 17. The application device according to claim 16, wherein one or more components of the application device, in particular gas sensing, pressure sensing, gas analysis, IR spectrometer, SEM or TEM devices, are used to manufacture an application device comprising each radiator device (1). And a second manufacturing process of manufacturing aligned with each of the radiator devices (1).

Images